December 2003 - cypherpunks-legacy

Even more unix holy war. Was "Clinton freezes U.S. assets .."
by jamesd＠com.informix.com 17 Dec '03

17 Dec '03

To summarize this thread, and I hope end it. The fans of unix correctly point out that unix can, in principle, be configured to be really nice. Well so can a molten lump of pig iron, in principle, be configured to be something really nice. Out in the real world people do not like dealing with a molten lump of pig iron, and so it is usually not configured to be really nice, which is why we see people editing with vi and publishing with latex. Now sometimes MIS creates a tiny little reserve where things really are really nice, and the secretaries are herded into this little reserve, and within this little reserve everything is almost as nice as windows. (Unless of course you want to cut and paste formatted text and graphics between windows, under the sun OS, but hey, that is an arcane operation that the normal user would never want to do. Right? You cannot really expect MIS to help such special power users can you?) And that is what Bill O'Hanlon meant when he said that for ordinary naive users who had not been exposed to either operating system, unix was no worse than dos. Herd them into the reserve and keep them there, and it is indeed no worse than dos. But the reserve is not really the whole operating system, it is just a little area of safety. And as soon as somebody strolls outside the little reserve to do something that MIS has not specifically planned for and arranged, half a ton of molten pig iron pours down on them, and MIS says: "Hey, don't do that." --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.catalog.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the omnipotent state. | jamesd(a)netcom.com

1 0

Re: white noise remailer
by Jaeson.M.Engle＠josaiah.sewanee.edu 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- This is an interesting approach. It could even contain a small set of PGP'd messages like "This is the sound of white noise on the Internet" and "Have fun breaking this one?" In the even someone took the time to break the pgp... >I'm new to all this so please pardon my ignorance, but are you guys >consdidering the following approach: > >The remailer transmits a constant bandwidth of outgoing messages of all >about the same size to a uniformly distributed range of target addresses. >Incoming messages are randomly dispersed within this stream, packed out to >the standard size. > >-rkm > > >--- > >Richard K. Moore - rkmoore(a)iol.ie - Wexford, Ireland - fax +353 53 23970 -----BEGIN PGP SIGNATURE----- Version: 2.6ui iQCVAgUBLyhlIUiLvmPjc4XdAQFRgwQAiSJr09cQav2c+kqz+6a8NSyh4BXf1prx 0hDbn31MlBBWTnezk/+PSCmfE9eXZwLbrLDCtUwIoudhIYxTqAzJV10GK0gWyF7O 6iKg6LTEgsa4DfJC2uc7Knrdj9ln9LK6qy8/I5zM5QZdfZHwYEuEcGS5aPIHdYGT SbKsnUi0rHI= =kUQl -----END PGP SIGNATURE----- aka: (-: Jaeson M. Engle || jme(a)josaiah.sewanee.edu :-) (-: www server: http://josaiah.sewanee.edu/ :-) (-: It's February 8th! IT'S TIME!!! Ask me for details!:-) (-: Finger 'jme(a)josaiah.sewanee.edu' for my Public :-) PGP block.

1 0

Even more unix holy war. Was "Clinton freezes U.S. assets .."
by jamesd＠com.informix.com 17 Dec '03

17 Dec '03

I am writing this under sun's unix imitation of Microsoft's and Apples GUI. I am using their almost acceptable ASCII editor. I leave a window open for a long time. Then I attempt close down the window, an operation that I can do in Microsoft Windows with a single double click. In unix I click on the border to bring to popup a menu, carefull drag to the right point on the menu, then release. In addition the menu names are confusing. If you want to minimize a window you have to "close" it. If you want to cause a window to cease to be, you have to "quit" it -- Hey did I not already quit it when I forgot about it and went to do something else. A warning message comes up: I have not saved. I am offered the options of quiting or not quitting. Gee guys, did it never occur to the great unix gurus that at this point the option that the user most likely wants to do is save the file. How come the number one most common option is not in the message box. A thought strikes me. I have set up Microsoft windows with a little icon that allows me to reboot or close down windows with a single double click. What will happen if I attempt to reboot unix. Well the answer is obvious, seeing as unix is made of broken glass with sharp edges everywhere, working on the principle of "You asked for it sucker, you got it, Ha Ha, massive data loss to prove you are no good at unix." So I go to a terminal emulation box (Notice how under unix windows the number running app is terminal emulator, whereas in Microsoft windows people use Windows applications.) I make myself root, and enter my root password. I type sync three times, and then type reboot. And lo, unix reboots. Note that this is substantially more work than double clicking on an icon. Note also that rebooting takes *much* longer than with DOS/Windows. And, as can be predicted from unix basic operating philosophy (Cut the stupid users throat if you can) my open edit windows are cut down *without* any message boxes popping up giving me the option of saving this file before the reboot proceeds. Eventually, at long, long, long, long, long, last, Unix deigns to come back and give me some windows. Naturally the test edits that I did just before rebooting are lost, as I expected from the basic unix cut-the-stupid-suckers-throat philosophy. What I did not expect is that one of the files I *did* want (and saved immediately before rebooting), has not only not kept the changes I made -- the file has completely DISAPPEARED! -- Too bad sucker -- that will teach me -- maybe next time I will type six sync commands before rebooting as a sacrifice to the great God unix in the vain hope that it will spare my data. Or perhaps no number of sync commands could make a difference because some network hiccup postponed the file save, (I was saving to a network drive) so that the file only *appeared* to be saved, but was in reality somewhere on my machine, or somewhere on the network, in unsynced limbo, and my privileges are not such that I can sync the network. Or perhaps it is just another of the many profound mysteries of unix that enable unix gurus to make a living. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.catalog.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the omnipotent state. | jamesd(a)netcom.com

2 1

Even more unix holy war. Was "Clinton freezes U.S. assets .."
by jamesd＠com.informix.com 17 Dec '03

17 Dec '03

And did I mention that "make" used to treat a tab as semanticly different from a space, so that a tab with an invisible space in front of it would result in the makefile failing. Oh yes, I did mention that. Well I guess I will mention it again. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.catalog.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the omnipotent state. | jamesd(a)netcom.com

2 1

Re: LOCKSMITH'S GUILD WANTS L
by doumakes＠netcom.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- "James A. Donald" <jamesd(a)netcom.com> wrote: > This assumes that governments have the necessary wisdom to tell a bad > architect from a good architect. This does not seem very plausible. > > [snip] > > While universities often issue or refrain from issuing certificates for > corrupt and improper reasons, government supervision on this matter is > unlikely to keep them honest I think you misunderstand the licensing process. The standards for a profession are set by the profession itself. They decide what is an adequate curriculum, for example, and what is an adequate amount of continuing education. They literally write the questions for the licensing exam. They also privately accredit the schools, which is more than adequate to "keep them honest." Then the government administers the exam and issues licenses. The professional associations could do this themselves, making it an entirely private affair, but there remains one function that only government can exert: the government forbids someone without a license from functioning as a professional. In other words, licensing isn't a process of government interference in professions. It's a process of professions seeking governmental imprimateur for their own credentialing process, arguably in the public interest. (YMMV; there may be exceptional professions that just hated the whole idea and were dragged into it kicking and screaming, of which I am unaware.) ObCrypto: It's certainly true that crypto will make it easier to function as a professional without having the credentials. "On the Internet, nobody knows you're a quack." I just see it as something to endure, rather than something to applaud. ______________________________________________________________________ Don Doumakes Finger doumakes(a)netcom.com for PGP public key Foxpro databases built to your specifications. Email me for details. -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBLygTahtumcu2AjihAQEb7AP8DRfcZJvLZy8mw7jg6wjrjHeNl1IbP5EW h3en/84ENKB2SbmcIXx7TtEl3jGz6bZ2T7JA7Ed3UYC78lq8/QsTgBW41Sh+arbX /j1o6Sot6jYAs5EGiuiD4OoxhRopDMuoZFt9NyRJfTF68jHP1olsjiLwnV0mq3g4 nRUh71wnFmk= =EjEG -----END PGP SIGNATURE-----

2 1

Even more unix holy war. Was "Clinton freezes U.S. assets .."
by jamesd＠com.informix.com 17 Dec '03

17 Dec '03

And did I mention that in DOS you can type rename *.txt *.asc to turn all *.txt files into the corresponding *.asc file but that if in unix you type mv *.txt *.asc not only will it not rename your *.txt files to be *.asc files, but if you have only two *.txt files it will silently destroy the second and issue no warning that anything has gone amiss. Hell, it is only the users data -- stupid user, he can type it in again. And of course I have already mentioned the unix "make" utility. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.catalog.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the omnipotent state. | jamesd(a)netcom.com

3 3

Even more unix holy war. Was "Clinton freezes U.S. assets .."
by jamesd＠com.informix.com 17 Dec '03

17 Dec '03

And did I mention that in DOS you can type: dir filename /s if you suspect that filename is in a subdirectory somewhere and you want the computer to find it for you, but in unix you type: find . -name filename -print but the entertaining thing is that if you type find . -name filename find then merrily goes forth, searches for the file -- which takes an extraordinarily long time, much longer than on my DOS machine, and then, cheerfully throws away what it has found, no warnings, no error message, no explanation. find thinks to itself "Gee, the user asked me to find the file, but he did not tell me to do anything with the information, so, I guess he just wanted me to exercise the network and the hard drive for five minutes or so." And did I mention "curses" You are spared. I will not mention "curses". And did I mention SUID. Well fortunately SUID has already been mentioned by other folk in the "Clinton freezes .." thread. And did I mention that unix has no less than seven families of metasyntactic operators, each of them different from and clashing with the others, that these families follow no logical order of substitution that I can figure out. Yeah, I know, that is what makes the unix shell so powerful. It is remarkably similar to what makes a chainsaw so powerful. It is very good at cutting your fingers off. And once again, I wish to remind you that the unix "make" utility used to treat spaces as semanticly different from tabs. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.catalog.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the omnipotent state. | jamesd(a)netcom.com

3 2

anon communication
by Harry S. Hawk 17 Dec '03

17 Dec '03

Re: anon communications Hello... There are many reason that anonymous communications are needed, I will just focus briefly on one that is important to our business. When we post jobs on the Internet we need to remain anonymous. We do this when we place ads in the NY Times as well. This is just one reason why anonymous communication on the Internet is very important. Harry Hawk Manager of Computer Services Warwick Baker & Fiore 212 941 4438 habs(a)warwick.com /hawk

1 0

Even more unix holy war. Was "Clinton freezes U.S. assets .."
by jamesd＠com.informix.com 17 Dec '03

17 Dec '03

And another vast advantage of DOS Windows is that it is a vastly better environment for developing software than unix. The great strength is of course symbolic debugging -- you can single step your compiled code, and see it displayed symbolicly, with the symbols and statements of your source code, whilst unix programmers are usually reduced to picking through core dumps like grave robbers. You can run your program under the debugger and set it so that when an exception condition occurs, bingo you are in the debugger at the line where the exception occurred, and all the variables as they were when the exception occurred, and the rest of windows, the graphical user interface, has been frozen until you examine the situation to your satisfaction. Try that in Unix. You might be able to do a crude and limited equivalent with a text mode program, but with a GUI program you are hosed. This is the basic reason why unix software sucks -- because the tools for writing it suck. And every unix programmer gets to roll his own user interface, which means that every programmer has to figure out how to roll his own user interface and every programmer does it differently. And windows has the *.rc file system for internationalization. A simple and powerful method for putting everything that needs to be translated in one set of files, and nothing that needs to be translated in the other set. Unix has no equivalent. So every company that needs to internationalize its product hand rolls their own ragged equivalent to to windows *.rc and *.res files, but it cannot be the same because they never hand roll the tools for manipulating and using *.rc and *.res files. Windows has power tools for dealing with this problem. Unix does not. And the power tools that unix does possess not only do not adequately address this problem, they are also likely to cut your hands off. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.catalog.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the omnipotent state. | jamesd(a)netcom.com

3 2

Re: Reordering, not Latency (Was: Re: Remailer)
by nobody＠replay.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- On Wed, 25 Jan 1995 Michael Handler wrote: >On Wed, 25 Jan 1995 Louis Cypher wrote: > >> In recent discussions, the consensus >> was that message reordering was superior to (and the actual intent of) >> latency. Reordering is not sufficient, a form of latency is required >> to make it effective. > > I have literally hundreds of messages archived from the CP list of >several months back where Eric Hughes repeatedly states that reordering, >not latency, is the key. Reordering of a sufficient magnitude will >introduce latency inherently. Otherwise you are still vulnerable to >traffic analysis (which is an art, not a science, remember). > >-- >Michael Handler <grendel(a)netaxs.com> If you read my analysis with more care, you will note that I do not argue that latency is superior to reordering, nor that it should be used in place of reordering. Reordering is far superior to latency. My point is that pure reordering is vulnerable to attack. It is less vulnerable if combined with latency. Also note that this was a repost of a message I sent to the list some months ago. I sent it because it seemed similar to the work that Wei Dai has been posting. -Louis Cypher -----BEGIN PGP SIGNATURE----- Version: 2.6 iQCVAwUBLyiy0qyHUAO76TvRAQF6qAQAk6IDl+UknPQ+c6hbmPvlgCtF2xwG0KZ3 aUXiuv/XJ5e5prRwLQ4X6rOUSflT2f+mRPO4fveVmWePWcr8BtV6LUQM+qJOLe5k Ay7CD+OGN9ni9EhRbIg3Sgdv6yB8cX2CpdMSByFD8J9rM240UWqSt/DgNQIlkTtC Jx+0NQP80QA= =aKOE -----END PGP SIGNATURE-----

1 0